Neural Networks are becoming more and more valuable tools for system modeli
ng and function approximation as computing power of microcomputers increase
. Modeling of complex ecological systems such as reservoir limnology is ver
y difficult since the ecological interactions within a reservoir are diffic
ult to define mathematically and are usually system specific. To illustrate
the potential use of Neural Networks in ecological modeling, a software wa
s developed to train the data from Keban Dam Reservoir by backpropogation a
lgorithm. Although the available data was insufficient and irregular, the s
ystem was trained successfully to estimate the chlorophyll-a concentration
given the time, total suspended solids, total phosphorus, dissolved inorgan
ic nitrogen and secchi depth. The model was quite successful in estimating
the output with an average error of 0.01268 to 8.11612x10(-8) percent for t
he 5 sampling stations.